def test_transform_range(self):
columns = ["Product number", "Quantity", "Retailer code"]
data = self.go_sales[0][columns]
data_spark = lale.datasets.pandas2spark(data)
sk_scaler = SkMinMaxScaler(feature_range=(-5, 5))
rasl_scaler = RaslMinMaxScaler(feature_range=(-5, 5))
sk_trained = sk_scaler.fit(data)
rasl_trained = rasl_scaler.fit(data_spark)
sk_transformed = sk_trained.transform(data)
rasl_transformed = rasl_trained.transform(data_spark)
rasl_transformed = rasl_transformed.toPandas()
self.assertAlmostEqual(sk_transformed[0, 0], rasl_transformed.iloc[0,
0])
self.assertAlmostEqual(sk_transformed[0, 1], rasl_transformed.iloc[0,
1])
self.assertAlmostEqual(sk_transformed[0, 2], rasl_transformed.iloc[0,
2])
self.assertAlmostEqual(sk_transformed[10, 0], rasl_transformed.iloc[10,
0])
self.assertAlmostEqual(sk_transformed[10, 1], rasl_transformed.iloc[10,
1])
self.assertAlmostEqual(sk_transformed[10, 2], rasl_transformed.iloc[10,
2])
self.assertAlmostEqual(sk_transformed[20, 0], rasl_transformed.iloc[20,
0])
self.assertAlmostEqual(sk_transformed[20, 1], rasl_transformed.iloc[20,
1])
self.assertAlmostEqual(sk_transformed[20, 2], rasl_transformed.iloc[20,
2])
def test_transform(self):
columns = ["Product number", "Quantity", "Retailer code"]
data = self.go_sales[0][columns]
sk_scaler = SkMinMaxScaler()
rasl_scaler = RaslMinMaxScaler()
sk_trained = sk_scaler.fit(data)
rasl_trained = rasl_scaler.fit(data)
sk_transformed = sk_trained.transform(data)
rasl_transformed = rasl_trained.transform(data)
self.assertAlmostEqual(sk_transformed[0, 0], rasl_transformed.iloc[0,
0])
self.assertAlmostEqual(sk_transformed[0, 1], rasl_transformed.iloc[0,
1])
self.assertAlmostEqual(sk_transformed[0, 2], rasl_transformed.iloc[0,
2])
self.assertAlmostEqual(sk_transformed[10, 0], rasl_transformed.iloc[10,
0])
self.assertAlmostEqual(sk_transformed[10, 1], rasl_transformed.iloc[10,
1])
self.assertAlmostEqual(sk_transformed[10, 2], rasl_transformed.iloc[10,
2])
self.assertAlmostEqual(sk_transformed[20, 0], rasl_transformed.iloc[20,
0])
self.assertAlmostEqual(sk_transformed[20, 1], rasl_transformed.iloc[20,
1])
self.assertAlmostEqual(sk_transformed[20, 2], rasl_transformed.iloc[20,
2])
def test_fit(self):
columns = ["Product number", "Quantity", "Retailer code"]
data = self.go_sales[0][columns]
sk_scaler = SkMinMaxScaler()
rasl_scaler = RaslMinMaxScaler()
sk_trained = sk_scaler.fit(data)
rasl_trained = rasl_scaler.fit(data)
self._check_trained(sk_trained, rasl_trained)
def test_fit_range(self):
columns = ["Product number", "Quantity", "Retailer code"]
data = self.go_sales[0][columns]
data_spark = lale.datasets.pandas2spark(data)
sk_scaler = SkMinMaxScaler(feature_range=(-5, 5))
rasl_scaler = RaslMinMaxScaler(feature_range=(-5, 5))
sk_trained = sk_scaler.fit(data)
rasl_trained = rasl_scaler.fit(data_spark)
self._check_trained(sk_trained, rasl_trained)
def test_partial_fit(self):
columns = ["Product number", "Quantity", "Retailer code"]
data = self.go_sales[0][columns]
data1 = data[:10]
data2 = data[10:100]
data3 = data[100:]
sk_scaler = SkMinMaxScaler()
rasl_scaler = RaslMinMaxScaler()
sk_trained = sk_scaler.partial_fit(data1)
rasl_trained = rasl_scaler.partial_fit(data1)
self._check_trained(sk_trained, rasl_trained)
sk_trained = sk_scaler.partial_fit(data2)
rasl_trained = rasl_scaler.partial_fit(data2)
self._check_trained(sk_trained, rasl_trained)
sk_trained = sk_scaler.partial_fit(data3)
rasl_trained = rasl_scaler.partial_fit(data3)
self._check_trained(sk_trained, rasl_trained)
def test_get_params(self):
sk_scaler = SkMinMaxScaler()
rasl_scaler = RaslMinMaxScaler()
sk_params = sk_scaler.get_params()
rasl_params = rasl_scaler.get_params()
self.assertDictContainsSubset(sk_params, rasl_params)
def test_pipeline_pandas(self):
pipeline = RaslMinMaxScaler() >> LogisticRegression()
trained = pipeline.fit(self.X_train, self.y_train)
_ = trained.predict(self.X_test)